Dr. Gareth Morgan, MD, FRCP, FRCPath, PhD Dr. Christoph Heuck, MD Dr. Niels Weinhold, PhD Dr. Jacob Epstein, DSc Dr. Shmuel Yaccoby, PhD University of Arkansas, Myeloma Institute for Research Therapy Interview Date: May 18, 2015 Summary You wouldn't think that having new drugs in the myeloma pipeline would actually be a challenge, but with an increase in new treatments also comes THE big question: Which therapy should each myeloma patient receive? Creating a clinical trial for each type of myeloma patient and each possible therapy combination could take decades. Drs. Morgan, Heuck, Weinhold, Epstein and Yaccoby propose the use of a mouse avatar that has your exact kind of myeloma. With this avatar, they can test how a drug or combination of drugs could act before the drugs are used in live patients, determining how effective they would truly be and shaving years off of important discoveries. The Myeloma Crowd Radio Show with Drs. Morgan, Heuck, Weinhold, Epstein and Yaccoby.
Jenny: Welcome to today's episode of Myeloma Crowd Radio, a show that connects patients with myeloma researchers. I'm your host Jenny Ahlstrom and I'm joined today by my myeloma friends including Gary Petersen, Jack Aiello, Pat Killingsworth and Cindy Chmielewski as co-hosts. This is the eight and a very important series featuring the Myeloma Crowd Research Initiative. We believe that patients can help find and fund curative research for myeloma. And you might wonder how we do this. We first created an expert scientific advisory board, then a patient activist advisory board. We decided to go after solutions for high-risk myeloma because these patients have dismal outcomes even with the advances that have been made. We also believe that by addressing high-risk, the new discoveries will ultimately help all patients because we all become high-risk at some point in our myeloma journey. We called for letters of intent and receive back 36 high quality proposals from top investigators around the world. Our scientific advisory board then scored these proposals and selected a top ten. This show is number eight in the ten. And after the shows are complete and a full proposal of these ten have been submitted, both a scientific advisory board and the patient advisory board will vote select a small number of proposals to fund. We would invite you to become part of this initiative and encourage you to listen, or read the shows, share them with your myeloma friends so that you can help us decide what is the most critical research to move the bar in myeloma treatments, especially for patients with these high risk and very aggressive features. We would love to hear from you and want you to be involved because this is a team effort. So we are very privileged today to have many doctors and researchers from UAMS with us today including Dr. Gareth Morgan, Dr. Niels Weinhold, Dr. Christoph Heuck, and Dr. Joshupa Epstein, and Dr. Shmuel Yaccoby. So welcome, doctors.
Jenny: Thank you very much for joining us. I didn't know all of you were joining, but I have an introduction for three of you I know. Dr. Gareth Morgan is director of the myeloma institute at UAMS, deputy director of the Winthrop P. Rockefeller Cancer institute at UAMS and Professor of Medicine. Dr. Morgan came to UAMS from the Royal Marsden NHS foundation Trust and The Institute of Cancer Research in London, Europe's largest comprehensive cancer institute where he was a Professor of Haematology and Director of the Centre for Myeloma Research. He's also founding director of the European Myeloma Network and has authored more than 45- peer-reviewed journals. Dr. Niels Weinhold is a Research Assistant Professor at UAMS. His field of study includes the genetics of multiple myeloma, minimal residual disease, and diagnostics in multiple myeloma. Prior to his coming to UAMS he was a Postdoctoral Fellow in the Department of Internal Medicine at the University of Heidelberg, in Germany. Dr. Heuck is Assistant Professor of Medicine at the Myeloma Institute for Research & Therapy. He completed a hematology/oncology fellowship at Montefiore Medical Center and an Internal Medicine residency both in New York. Dr. Heuck is board certified and has conducted research related to molecular genetics of myeloma and other cancers. So all of you, welcome. Thank you so much for joining us today. We're very appreciative.
Doctors: Thanks again.
Jenny: So let's begin. Cindy is on the line as well. So let's begin with the proposal that you submitted. It's more related to the diagnostics and trying to discover why these clones are so aggressive, what makes them aggressive, what causes them potentially to be aggressive and use that to determine appropriate therapy strategies for high-risk myeloma patients. So in your proposal you said there are typically five sub-clones present at presentations. And we've asked a lot trying to dig into the different genetic features. Are these genetic features? Are these the translocations that you hear about like 4;14 and 11;14, and deletion 17, or are these different sub-clone types?
Dr. Morgan: It's really an important question to ask. So the knowledge of cancer has undergone quite a change in the last two to three years. So in the past, the idea was that cancer was clonal which meant everything in the cancer cell was exactly the same so they all came from a single stem cell and looked and behaved the same. What’s become clear is that that's not true, that when you look at different sites within a tumor, that they can be subtly different from one another. And so the idea is within the cancer as a whole, there's a mixture of subtly different stem cells that have different behaviors, some with more aggressive and some with less aggressive behavior patterns. And understanding those behavior patterns and targeting the treatment to the stem cell that causes the myeloma is really what we should be doing. And at one level, this whole project is designed around that concept.
Jenny: Okay and it looks like you are working using whole exome sequencing. Maybe for those who are not familiar with that type of testing, could you just give a quick overview of what that is any why are you using that instead of, let's say a FISH panel or a gene expression profile, or something.
Dr. Morgan: So the important thing about the sequencing and technology is that you don't just look at one location in a tumor or in a gene to look for a mutation. You can look at all of the genes that are present in the cells. So largely, you gain much, much more information. FISH looks at one specific point. When you use a sequencing panel, you can look at all of the genes that make up the human body so you get just so much more information. And you can also use that information to work out what percentage of cells actually have that mutation and define the sub-clonal structure of the cancerous cells. So it's much more powerful than we were in the past.
Jenny: So when you're saying you have five sub-clones present at presentation, you're saying that on average, the typical myeloma patient has five different types of myeloma in their myeloma cells at one time, when they're really diagnosed. Is that correct?
Dr. Morgan: Absolutely. And each one of those cells we have some different characteristics. So it's possible to have a small population when you first present that has high risk characteristics and over time, that cell can grow out, come to dominate everything and lead to relapse. And if you go back a few years, this concept of clonal tiding was developed where you can have a relapse, develop high-risk disease and at the end of the disease journey - and that sub-clone might not have been obvious at any time during the whole natural history of the disease. So we need special testing to identify them early on to be able to do something therapeutically about them.
Jenny: And so the whole exome sequencing identifies what those five clones are so then you know what you're working with.
Dr. Morgan: Yes, absolutely. It's a cool technique for doing that.
Jenny: And let me just ask before we dig in to all the details, are you considering to be high risk for your study? Are you looking just at genetic features? Are you looking at late stage? Are you looking for the cells that are chemo resistant? So maybe you could clarify for us.
Dr. Morgan: I think it’s a real challenge for you. You're having an initiative to deal with high risk. And we kind of all know what we mean by high risk, but there's no consistent definition. So what we may mean by high risk in Little Rock might be entirely different from what somebody in New York is considering high risk. And so it's about primary treatment resistance. It's about using the gene expression to define the GEP70 high-risk group. We know those behave really badly with treatment and it's a very good test for identifying people with high risk. The FISH technology, so just using a single FISH probe and saying there's a high risk translocation really doesn't work because within a translocation group, that could be good risk, poor risk, intermediate risk. So the definitions of high-risk really need to be pushed forward and there needs to be a consensus reached.
Jenny: Yeah I would agree. I think that's probably where you're all headed next I would guess.
Dr. Morgan: Yes. I think that's one of the most important things we can do.
Dr. Heuck: This is Chris Heuck. So but for the purpose of this study, what we're really looking at is the patients who have a high risk defined by the gene expression profile. Dr.Barlogie over the last 15-20 years had done tremendous amount of work where we identified the signature that really tells us that this about 15% of all newly diagnosed patients have this high-risk behavior. And so this is going to be our selective for this study.
Jenny: Okay. It is tricky trying then to find who and what is high-risk. But let's start in with your findings a little bit. You were saying in your preliminary data that you found the early relapse and shorter overall survival were associated with high-clonal diversity. So maybe the more types of myeloma you have, the more risk you have in having one of those clones grow up and be highly aggressive. Do you want to discuss that at all?
Dr. Morgan: These concept developed from ecology and relate to the idea that the more diverse something is, the more variation there is in it, the more likelihood that one of those variants is going to be associated with high-risk. So in many respects, if you just got one clone and it sits there, it's not dividing and stops other clones growing out, that's really good risk because there's just one genetic feature. If you have lots of genetic features that within that, it's almost certain that they'll be something that's more high-risk that can grow out of some stage during the disease. And at one level we still need to work out whether aggressive, early intervention for people that have relatively low risk disease may or may not be the right way forward. And this type of experiment can now answer that as a question.
Jenny: And that's why -- is it CLL where Gleevec is used because there's one mutation in all the cells and there's one target basically?
Dr. Morgan: Yes, basically. That's why chronic myeloid leukemia a curable disease. It's a very simple type of cancer that probably has just two mutations. And one of them can be targeted with the drug. And so that's why we make so much progress so rapidly with CML. But we needed different approach in myeloma. And the joy of this approach is putting together the genetics with an animal model and almost what we're talking about this developing an avatar which is a groovy word that describes a way that you can model an individual patient in a mouse model. And then you can test treatment, you can test its genetics without having to really put the patients at risk. You just model in the animal. So you have an avatar of that disease which you can treat in different ways. And so that's really where we're trying to go with this. What do you think?
Jenny: Oh, well, I think it's -- yeah, go ahead.
Dr. Heuck: Gareth, said that's exactly what we need to go and try to be able to model the individual patients in order to find the most fitting therapy for those patients without exposing them to unnecessary toxicity in the future.
Dr. Morgan: Because about what we think is that the number of treatments is going to explode. And that we'll have more and more treatments, and the challenge is picking the right treatment for the patient. And while the genetics has been helpful, somehow understanding what happens in model systems like these mouse avatars are really going to push the whole field forward.
Jenny: Well, I haven't heard of the mouse avatars before. Is that fairly new?
Dr. Morgan: We hope so.
Jenny: Oh, okay.
Dr. Morgan: It's an idea that's out there. It's causing a lot of excitement because evaluating new treatments in model systems is not been as good as it might be up to now. And the system that Shmuel and Josh have developed here where you can grow high-risk myeloma in a mouse avatar, and then use those avatars to test treatment, really strikes me as being the most exciting challenge in myeloma today, and certainly in high-risk.
Jenny: Well, that would be an amazing way to allow very personalized therapy to say okay, We're going to create a mouse model that has 11;14 but has high LDH levels or something. I'm putting something crazy out there. So in your research, you said patients with a dominant clone greater 25% typically have delayed progression free survival, so and this clone might suppress other more potentially aggressive clones and not allow to them to grow up. But I had a follow up question with that. That is if you have a risk feature that is this dominant clone like let's say 4;14 or 14;16 or something, will you do better than a low or standard risk patients who have a variety of clones but no high risk features?
Dr. Morgan: So the important thing to say is that not all people with a 4;14 and not all people with a 14;16 that deregulates MAF behave badly, and it's -- one of these paradoxes that are out there which is why do some people with those lesions behave well and live for a very long time and they're eve cured compared to others with exactly the same molecular features who do quite so badly, and that's what this type of model system can allow you to address and what we're putting forward is one sort of hypothesis that may explain why there's a difference within different molecular sub-groups.
Jenny: That makes sense. Do you want to give us an overview for patients who haven't read the proposal about you are trying to accomplish and what you would like to determine through this study?
Dr. Morgan: So the real aim is to understand the biology of the stem cell that underlies high-risk behavior by taking patient material, growing them in different mice models, and then transplanting them from one mouse to another, and to see what happens to the genetics over time as a marker which will illuminate the stem cell that is underlying that whole sub-clone of myeloma because if we could understand that, then we'll be able to target treatment to that stem cell and really develop these mouse avatars where treatment can be evaluated in the mouse because we'll understand exactly how the individual's myeloma is composed.
Dr. Heuck: Another feature and you mentioned this before is that risk models are -- we can always have a better idea of risk in retrospect. If we thought a patient was low-risk but relapsed early on in therapy, he clearly declared him -- or he or she clearly declared himself as a high-risk patient, which also means that any of the predictive risk models are really highly dependent on the therapies that are given to the patient. And one feature that we're really interested in is finding out how different therapies drive certain clones to survive or which clonal features are associated with resistance to particular therapy. So one aim or one feature of this proposal will also be to use these avatars, treat them with a typical myeloma treatment which would be Velcade, nobody would give single agent Velcade to a patient nowadays but what it helps us to really understand is what does Velcade do to a particular high-risk clone, which sub-clones may survive, are there mutations that arise, are there pre-existing mutations that make one clone resistant and we can do that in this mouse model without exposing the patient to inadequate therapies.
Jenny: Okay well, let's talk about that for a minute because that's a very big deal to think about the stress of intensive treatment may actually cause a clone to become more resistant and aggressive. Can you address that -- how you determine that might be possible?
Dr. Morgan: So that conceptually, we've always gone with this idea that you should use multiple treatments that have cross cover of the cancer cells to try and get rid of many sub-clones at presentation as we can to decrease a likelihood that one cell remains and grows out later on. And I think that's been a good concept I think. The Total Therapy protocol has actually delivered. There of 30% more people that are actually cured long term, so the concept works. The question is how do we develop that further and why hasn't the Total Therapy protocol been quite as effective for high-risk disease as it has been for low-risk disease. And so what we're trying to do is to say what treatment combinations and strategies should we use in a high-risk patient? Should we avoid chemotherapy altogether? There's a possibility that the chemotherapy could enhance disease progression by opening up space for sub-clones which are more aggressive to live in. And so there's a great deal to be learned but unless we start doing it in these kind of models, we're never going to make any progress because we always do our best for patients. We should always choose the optimum treatment for them, deliver it in the best fashion. When you take the cells out of the patient and then model their behavior in a mouse, you can change these kind of treatment strategies to see what would've happened if you’d done it in the patient. And that's the exciting there that Chris was trying to explain and I’m not sure we are explaining it as well as we might but the idea is really very appealing to study patient disease behavior in a mouse avatar where you can use any treatment that you fancy and see what that does to the behavior of the cells in the mouse.
Jenny: Well I love that approach and I think we all do because with the number of new therapies that are out there, even trying to pick a clinical trial to join is a crap shoot for patients. "Do I go with an immunotherapy therapy? Do I go with a vaccine? Do I go with a Kinase Inhibitor?" You could list ten or 15 different options now so for you to be able to do that in the lab in mouse models and do it in a test environment to say for this specific kind of clone, we know that this will shut that down and then if you have this and this then these two together might work. That's game changing I think for myeloma therapy.
Dr. Morgan: Yes. Information is power for the myeloma patient, and understanding their disease, and how it's going to progress and respond to the different therapies. It's really a way of improving outcomes, reducing side effects, and making expensive pharmaceuticals more efficient and improving their value really.
Jenny: Alright. I have three other patient advisory board members on the line. And I know I have other questions but I want to give them an opportunity also. So Jack would you like to start with any questions that you have?
Jack: Sure. Thank you for having me. I think we've all seen the pie charts of -- let's say the green being the dominant color in a myeloma cell and as treatment goes on, the colors change and subsequently, the blue becomes the dominant color, and those are different clones right?
Dr. Morgan: Yes. Absolutely.
Jack: I guess my question is is the correlation between the gene expressions profiling for green? If I would do a subsequent gene expression profiling for the blue?
Dr. Morgan: Just because we get a bit pedantic about things so that's important for the understanding. So the gene expression profiling is looking at the genes that are present, that are being expressed in the cell. And that's this GEP70 which defines high-risk very well. Those genes don't necessarily drive the behavior of high-risk, they're markers of high-risk. And what we're talking about doing is identifying those high-risk patients and then looking at then looking at the mutation pattern in their DNA to find the genes which are really driving the abnormal behavior. Does that make sense to you, Jack?
Jack: So does that mean the GEP 70 profile, does it change as treatment goes on? I know that patients can -- as these clones develop, they can end up with different translocations that they had in the beginning. Does the gene profile change?
Dr. Morgan: We did a very interesting study comparing risk stages with the GP 70 presentation and relapse. And so some patients with low-risk will relapse as high-risk and some cases with high-risk will relapse as low-risk. And that's a very interesting finding and it's about 20% of high risk will come back as low-risk and arguably you've actually killed the high-risk clone totally, allowing the low-risk cells to grow out. And one of the things we're planning to do is to do pairs of presentation and relapse to look at mutations that maybe important in driving that whole process.
Jack: So then the answer to the question is yes the gene expression profile actually changes -- can change as treatment goes on?
Dr. Morgan: Absolutely.
Jenny: Okay Gary, would you like to go ahead with your question?
Gary: Sure, thanks. Dr. Morgan, I loved what your philosophy was for early diagnosis and …….that the 85% group before you ever see any bone damage or anything like that, they have with breast cancer, being able to screen and get to it before it goes any place. So I think that's a great thing and for those 15% obviously with high risk, you've got some ideas that look pretty good. One of the things that I saw was when the primary clone is over 25% or so that it potentially reduces the high-risk clones from becoming more dominant. So the question would be maybe is a control philosophy or you attack the cancer but don't try to eliminate it, an approach in the short term that might have some benefit.
Dr. Morgan: I think that's a really fascinating concept that we don't know the answer to it but some of these -- in smoldering myeloma, the outstanding question is do you treat them early and try to reduce the amount of cells at risk or because they are just sitting there and arguably preventing the growth of high-risk clones, do you just watch and wait? It's very dangerous using chemo early on sometimes. The development of these immune therapies -- so the Revlimid, pomalidomide, the enhancement with anti CD38, antibodies, elotuzumab, all of those PDL-1 treatments. Those are going to spell a major change in the approach to myeloma in the next decade. And understanding what group and where to place them, is going to be very important. And in one way, this experiment we're proposing in high-risk disease will allow us to address some of these questions. And Dr. Yaccoby here, we were talking earlier about models actually in mice but then developing systems in the laboratory in the test tube almost which modeled similar things where you can address the impact of antibodies is going to be really important as well, which is a spin off from this experiment. Do you want to comment on that?
Dr. Yaccoby: So in addition to the unique animal models that we established and allow us to study high-risk disease and how the high-risk myeloma cells behave how the sub-clones are selected, we also have unique culture systems that allow us to grow these high-risk patients in an environment that capitulate the bone marrow environment in patients and allow us to study these targets such as novel antibodies or a distinct combination therapy.
Dr. Morgan: So one of the things we haven't mentioned which is relevant to what Shmuel just said is the niche in the bone marrow where the myeloma cell grows because everything is not just about the genetics of the myeloma cell, it modifies its environment to make it favorable to grow and so the myeloma niche in the bone marrow is a really important thing that can be targeted therapeutically.
Dr. Yaccoby: And that's why we plan to study and if it's possible to study this is in our unique animal models, so we're moving into agriculture terms in which we study how the clone cultivates the bone marrow micro environment for their advantage and that's something that we can do in this unique experimental system.
Gary: One of the questions I asked in one of the previous presentations that we had was it seems like everything works in vitro, right? And I said well, but then you put it in the micro environment of the bone marrow, and it doesn't work so good because of what you just talked about. And then I said well, why don't just throw a little bone marrow in there? See what it happens in vitro as well. I say that in just but I really mean it
Dr. Morgan: It's really what we were talking about and trying to do. In the animal model, we put either rabbit or human bone marrow which has got human stromal cells in the human myeloma niche. And Shmuel can even grow myeloma in a 3D model in the test tube that does really recapitulate the impact of the microenvironment. So it's much more like the real thing because people have been curing mice for years. You don't want to cure mice, or cure mice's cell lines. You want to know what an individual's myeloma behaves like, what it's doing to the stromal microenvironment. And how you can change it.
Dr. Yaccoby: And Gary, you have to remember that if we use bone marrow from patients, this bone marrow or the microenvironment is already manipulated or have been manipulated for many years in the patient and here in these models, we can actually purify this aggressive high-risk clone and see how it cultivates or interacts with the unmanipulated microenvironment and compare this to other patients. So it is true that we can use the patient system or the patient microenvironment to grow these cells but for understanding how these interactions involve, it is important to have a control system.
Gary: People always say they ‘ve done it on mouse model and you call this an avatar. So what's the difference between somebody's Velcade resistant mouse model and in your avatar other than maybe yours is much larger and blue? Dr. Morgan: Oh, no. So it's neither large nor blue sadly. The idea is that you're actually making a model of the patient. It's not a mouse model per se that is just about the mouse myeloma where you make the mouse develop. You use the mouse as a system where you can generate a model of the individual patient in the correct environment. We've always been able to cure mice models, you inject cell lines into their skin, you inject them into their femurs. They're not natural models for myelomas but this idea of an avatar is more around creating a human disease that's representative of a patient in a mouse model. And that's why you would call it an avatar because you're modelling themselves in a different environment where you can change the selective pressures in terms of treatment when they're exerted on them where you couldn't do that in the patient because the aim in the patient is always get them into the best remission you can, get rid of all of the cells, and cure people if you can. In this, you have a flexibility to do science experiments, real experiments with treatment on the patient's cells in a different system altogether where the patients are not at risk. And so it's a real aim of modern cancer discovery, really, to be able to make these models of individual patients. Gary: It's remarkable, it's outstanding, it's amazing what you're trying to do. The one last -- because I guess I lied - one last question before I go is that does this new avatar duplicate what's done in an avatar, what does it duplicate, what happens in patient, I guess you can't answer that yet because that's always issue because what you do in a mouse model never seems to be able to be duplicated in the human and how certain are you that something like this can be accomplished with your avatar? Dr. Morgan: So I should really get Shmuel to talk about this because he's the man who's into agriculture and growing things. But between him and Josh they've set up this really nice system and we're now in a place where we can do some of this, so Josh. Dr. Epstein: Your question is well taken. Let me give you examples. One of the issues is myeloma is you’re really aware of is the bone disease. We’ve clearly demonstrated and reported this that what blocks bone disease in the mouse model is also very effective in patients. As is true for most model systems be it ours or anybody else's, it's much more effective in detecting resistance to therapy then response to therapy or sensitivity mostly because there are other factors that are involved, it's not the cells' response alone. You have the whole microenvironment and you have the whole metabolism is different often between mice and patients, but if the mouse does not respond to therapy, you know that the patient will not. So in this regard there is quite a bit of deduction from the mouse toward the patient. Dr. Heuck: Maybe I can add on something to this. We actually have had situations where a high-risk patient had a mouse avatar and we've treated the mouse avatar with antibody-based therapy which showed to be effective, the patient did get that anti body then in real life and we saw an absolutely incredible response to that therapy. Unfortunately that patient has died in the meantime of other causes but that was really our first example where we see that this model does work and we can obviously we try to expand that to other drugs in the near future. Dr. Yaccoby: When we tested this specific antibody of biological drug against other avatar cases we didn't see response and that would give a type of recommendation for the clinical team and indeed and in some of these cases in which we could not see clear response in the mice system. We had a very in-depth argument whether to use these antibodies in patients. Dr. Morgan: So I think the bottom line of that discussion was that in preliminary work in a limited number of cases it looks as if the whole system will actually work and predict for response in a patient but we need to do more work now to be secure in that information because you wouldn't go straight into DNA. You need the information to be sure about the decisions that you take. And now is a good time to do this. Gary: Okay. Thank you very much. Jenny: Okay Pat, did you have questions? Pat: Yes, hi can you hear me? Jenny: Yep. Pat: It's all wonderful. I'm discouraged though. Gary would know a little about this for years. As patient advocates, we've been trying to help get the word out about Total Therapy and the research that you're doing there at MIRT and I'm a Mayo Clinic alumni, so it's a really a different approach and when I was diagnosed eight years ago, I was never given an option for something that even resembled total therapy, and I've got to tell you unfortunately, if you went to Dana Farber, or Mayo Clinic, or MD Anderson, I'm not even talking about a medical oncologist because I don't know how the heck you ever get those people up to date. Total Therapy is still not one of the options offered to a newly diagnosed low-risk myeloma patient and so to me this is part of an even bigger issue which is you're doing all of this work and you're going back and you're doing genetic testing and then you fine tuning the genetic testing, and then you're doing some more genetic testing, and how do we get the word out to other myeloma specialists and medical oncologists for that matter about what to do and how to proceed and what works best and you'd think that would all get picked up, but this UAMS PR issue and all of that, maybe that blocked some of it but it's very discouraging. And if you got some thoughts, I think it’s time to launch a public relations campaign probably. Dr. Morgan: So I couldn't argue with anything that you just said there. I don't necessarily understand it myself. I've been here for nine months, we've looked at all of the data as carefully as I possibly can. I don't see any trickery or skullduggery, the outcome of low-risk patients treated with Total Therapy is really very special. The cure paper that appeared in Blood I think is meaningful and I think you're right, it's not really our role to preach that as a message, and it's really from like yourselves and patient organizations that are responsible for that sort of thing. What we're talking about here in terms of this crowd funding initiative is how do fund research and progress in high-risk myeloma where the outcome is really bad. What is the way that you identify it? How do you treat it? How do you select from 22 new treatments the treatment that's really going to work for an individual patient? And what we're proposing I think is a way of addressing that second question. Pat: Absolutely and I get it and I guess I'm -- I think maybe you're going to find that, maybe it's going to work. And then how do we get that message picked up? So that was just -- I know I was off topic just a little bit but yeah, I can't imagine that -- maybe it is, but lung cancer or some of these other cancers that it's so complicated that people go off in 20 different directions and we can’t pull them together. It's unfortunate. Dr. Morgan: One of the things about myeloma is you can learn a lot from myeloma about how to treat some of these other cancers. It's relatively simple. It's possible to understand it, the lessons that come from myeloma might translate to lung cancer and other cancers. So I think it's more to myeloma than is myeloma itself because it is such a good model for other things. Dr. Yaccoby: All of the information you asked about why don't our physicians use the Total Therapy? Information is there in great detail. It's imported to manuscripts and papers that all the physicians read, it's reported annually at least in two national meetings, It's a physician’s choice what they choose to offer patients. And the information is there, it's not our issue, public relations, but rather the recipients, the physicians. Dr. Morgan: But then the question was what we're going to do with the results of this study? Is it worth doing because everybody's going to ignore it? I think it's very clearly worth doing that. Jenny: I think based on what was Pat was saying, I think the question is for everyone is patients and researches want to say "Show me the data so I can make the best decision for my care." It's frustrating as a patient having a recommended therapy based on the doctor that you go to versus a standard protocol that says for my particular sub type, this would be the very best option. So I think that's what is frustrating for patients. Dr. Morgan: Yes it's frustrating for everybody because either really we're standing at that crossroads where we have to get more information and really try and personalize the treatment to a patient. Up until five years ago when there was -- you have melphalan, you have alkylating agents, iMiD drugs and proteasome inhibitors. We're really on the threshold now. I mean there are three drugs going to be approved this year; histone deacetylase inhibitors, anti-CD38s, elotuzumab, all of those kind of things. And once you have the set of treatments you can select from, you need the information to select the best. And really, that's what we're proposing is to generate these mice avatars, and then use them to evaluate different treatments on the patient's disease but in an avatar rather than in the patient. Jenny: And Cindy, are you on the line? Cynthia: Okay. I'm not quite sure if it was answered because of the audio trouble. I'm going to that clone issue again, five clones at presentation. And I'm trying to understand that. Are clones the same as mutations that you have in your gene expression or are clones different than those mutations that you may have as a deletion 17p and chromosome 13, or I'm just a little confused there? So could you explain what the clones are. Dr. Morgan: So think of it this way, when you look for mutations, sometimes they're only present in 30% of all of the cells. And so that tells you something is not all of the cells have that 17p minus. And it’s in a sub-population of cells that make up 30% of the total. So depending on how you treat that patient, if you treated them with something that didn't work against 17p, those 30% of cells would grow up and then they would finally make up 100% of all of the cells that are there. If you could target those cells with something that only kills 17p minus cells, then that's 30% would go to 0%. And so that's the basis of all of these arguments, that there are multiple different clones that are present. That it defines by their genetic makeup. And you should target the treatment to the genetic makeup of the individual cells. Cynthia: Great. Now these mouse avatars that we're talking about, in your vision would in the future of this work, a myeloma patient will have a mouse avatar that it based on their myeloma? Each patient will have their own mouse per se to see what treatments work best on it? Dr. Morgan: That's the idea is to really give everybody an avatar that you can do experiments on to be able to find the best treatments and it sounds very science-fictiony but it really it's the way forward, I think it can work. Dr. Yaccoby: Yeah I can tell you that we've done it in the past and one of the patients had -- he was very happy to see that he has more relatives around now. ? Jenny: Let me ask, how did you develop these mouse avatars? Has it been used before in work that you’ve done or in any other cancers? Or how did you determine that it would be an effective model to use. Dr. Morgan: Shmuel was really important in doing these experiments. I think the important thing he learned is not to inject the purified cancer cells, in this case the myeloma cells, it was to inject the whole bone marrow, because there are different components of the bone marrow, also engraft the transplanted bone marrow (bone chips) and then you do actually set up a system that is representative of what happens in the patient. It takes 6 weeks for them to grow and then you can take the cells from those bone chips and then you can inject them into a downstream mouse and another downstream mouse. That’s one of the definitions of what a stem cell is, is it’s capacity to repopulate different mice. I think that’s a significant breakthrough that really allows you to ask, what do the cells that keep the tumor alive in individual patients – how do they behave , what are their genetics, and how do they respond to treatment. Jenny: And then you are doing it in a more accurate environment. Dr. Morgan: Absolutely. It was always if you take a cell line and inject it under the skin of a mouse, it forms a tumor. If you give any treatment, the tumors don’t form and so people extrapolated from that that the treatments would be active when they clearly weren’t and there’s a couple of model systems out there that perform well. But really everybody’s tumors are so distinct that you need to see how that person’s tumor behaves rather than extrapolating from somebody else’s tumor to the individual patient that’s being treated. Jenny: I think it’s really a brilliant way to go about it. Now one of the other questions that we asked where I think we had some audio trouble was that we were asking if you were staring with Velcade. I think your short answer was yes, it’s probably the simplest place to start but do you want to elaborate on that a little? Dr. Morgan: Velcade is easy to use in the laboratory and we can use it as a way to see how different populations within the tumor respond to it. After we’ve done that a couple of times and learned the lessons from it, I think we want to be more expansive in our treatment selections. We really are seeing new treatments – the immune agents, the targeted treatments. Targeting the RAS pathway is really important to us and Christoph has done a lot with this MEK inhibitor drug, trametinib. Do you want to talk about that for a second? Dr. Heuck: Sure. So what was interesting and it came out of the need to find better therapies for patients with high-risk disease and who had seen anything, pretty much, that is the standard of care and is certainly not the standard of care in the rest of the world and we initiated a program where we did targeted sequencing of patients’ tumors and what came out of that was that we saw a lot of RAS mutations. We saw this in our data set and other people, especially Gareth who has done a lot of sequencing on newly diagnosed patients and saw the same thing, that the RAS/MAF kinase pathway is very frequently activated. Now the nice coincidence was that there was a drug that was recently approved for melanoma for patients that had a BRAF mutation. BRAF3600E and that targets a protein called MEK, which is basically just one step downstream from the RAS gene. Jenny: So this is MEK or MYC. Dr. Heuck: MEK. So it also turns out that BRAF is also just downstream of RAS, so the idea was well if RAS has a mutation and is activated, just blocking the pathway downstream using that MEK inhibitor should work. And because the drug was approved for clinical use, we were able to obtain it in an off-label fashion and we treated patients and were actually quite successful with some of them and we are very excited about this approach. Dr. Morgan: So the story is these MEK inhibitors work and they give you responses that are sometimes dramatic, even if you’ve had heavy-duty chemotherapy before and so they will likely work in high risk disease. Our conundrum is how do we know which patients are going to respond and how do we make the responses more durable. I think we can use these avatars as a way of answering that question. Jenny: Well, once you finish Velcade you could use daratumumab, and then you could use this MEK inhibitor, or all sorts of combinations. I guess that’s the challenge of having more therapies available – it’s a blessing but then it’s more complicated. How do you combine them and then you have dosing on top of that. Wow, that’s complex. Dr. Heuck: And as we go on, the more patients and the more tests we do, the more we’ll learn about how a patient’s genotype or genetic makeup will respond to certain therapy. So instead of having then to test a hundred different agents on a patient, we can already predict up front which agent will be most likely to have a good response and select those. Dr. Morgan: This whole precision medicine idea underlies where we are heading is that you should always choose the best agent for a patient that has the fewest side effects that has the highest chance of working. We need some tools to be able to deliver that idea. And one of the tools are just molecular diagnostic tests, but this would be the other side which is a more cellular system, a mouse avatar where you can really test what the molecular diagnostics predict. Jenny: Well, to follow up on that, Jack had some questions he wasn’t able to ask. He said, “Do you already have a list of predictive biomarkers on these clones, enabling drug development for those biomarkers. Dr. Morgan: We can get the biomarkers. I think the MEK inhibition and RAS mutation is a good example. So we know that if that pathway is activated, you will respond. But it could be that far more patients have activation of that pathway than our biomarkers predict. And it’s difficult to give somebody that doesn’t have a mutation an agent that targets that pathway. So you could test that idea in a mouse because you might not expect them to work, but on the other hand, if that pathway is activated because of the way the cells live together in the bone marrow then it may be that those are the ones that actually respond well to the treatment. It allows you to be testing ideas that you could never test in a patient. We have biomarkers but we need to be sure that the biomarkers are predictive of a clinical response we are interested in. Avatar mice would allow you to address those questions. Jenny: Yes, that you have some kind of solution. His other questions were around the genetic expression profile of the GEP70. Some patients know what that is and some patients don’t. Jack’s question is “Is the GEP70 precise enough to define high-risk for those 15% of patients?" Dr. Morgan: We have learned much from young Niels here. Niels and I have been working on this really heavily in the last few months so we should hear what Niels has to say. Dr. Weinhold: This is a question I may answer. The GEP70 was developed for those 15% of patients so there is technically a direct correlation according to the data we collected, the GEP70 is even more precise than published results based on FISH data. We think that the GEP70 is the best method for high-risk currently. Dr. Morgan: We did lots of work comparing the GEP70 to i-FISH, the interface FISH which people out there use as a standard and there is no doubt that the GEP70 was more accurate, shall we say, at identifying patients with high-risk behavior. It doesn’t identify all of the high-risk behavior, but if you are GEP70 positive, your clinical outcome is going to be aggressive and you are sure that people are going to relapse early and if you know that data, you alter your clinical approach to treatment – that is very, very clear and that is becoming more obvious. So it is good. It doesn’t identify all high-risk patients and in low risk there are some people with high-risk behavior, but if you are positive, you are sure that people are going to behave badly and need their treatment altered. Dr. Heuck: Just an argument people might make is, well the GEP70 was developed in Arkansas and only works in Arkansas and in Arkansas data, but there have been multiple validation studies using an external data set using data from the UK, data from Germany with other published, larger scale, even industry-sponsored trials and the GEP70 always came out on top in terms of identifying high-risk patients. Dr. Morgan: So it’s irrespective of treatment, that’s what Chris is saying. That’s always been an argument other sites have used, “Well, it only works for Arkansas patients.” I know it works for the British patients I treated in Britain. Niels tried it on the German dataset from Heidelberg and it works for those, and so it’s pretty clear that it’s a universally important test. Jenny: So here’s a follow up question. Why isn’t it used more frequently? I had a friend just recently diagnosed and I know enough now to be dangerous. I was saying, “They ran the FISH but did you ask them to run this Signal Genetics test?” And they said, “Well, ok” but I had to walk them through the whole thing about “You really need to do this before you start treatment, before you start killing all of your myeloma cells so you have some kind of marker as a baseline.” Why isn’t this test offered? Dr. Morgan: The test is available. It’s an impossible question that you ask us to answer because I don’t know why people don’t do that. All I can say is that our response to this has been, ok, so it doesn’t look like people do it. Maybe we should write a few papers that describe how this test behaves. So we’ve got two papers in preparation that we are going to send off in the next month or two that describe how the GEP70 works, how it compares to everything and why it should be important to people to know about the clinical behavior about people with GEP70 positive behavior. So those are going to go out . There’s also a movement in Europe to use gene expression profiling for the same approach. And I think it’s just now that GEP approaches are starting to get some traction and I think in the next year you’ll see more and more people using the GEP70 at other schools because they give such useful clinical information. Jenny: And I think it would be helpful for patients to know what the FISH provides, what the GEP70 provides and what whole exome or whole genome sequencing provides so that when they go to their doctor, they can request those and you start seeing a push from both directions. Dr. Morgan: Absolutely. I think that would really help. Jenny: Well we are happy to help describe that for patients so maybe we’ll talk about that later about how to get that information netted out for patients. Dr. Morgan: I’m tempted to put these papers, once we get them sent off, on our website so that people can see the data themselves. I think it is because people don’t know the data that they find it hard to believe. There is nearly 10 years of experience with it now and multiple validation sets. It’s kind of indisputable I think. Jenny: Well, you can promote it to the doctors and we can promote it to patients and maybe we’ll get some traction. Dr. Morgan: Good for you. Jenny: Jack also said, “Is there any way the GEP70 model could be integrated as part of the MRD testing or is it strictly used as a diagnostic tool? Dr. Morgan: You have to use it as a diagnostic tool really, but one of the interesting things which I found fascinating is that one of the subgroups has a totally distinct clinical course. There are two groups – one called CD1 and the other called CD2. It makes not a jot of difference what those groups are, other than the CD2 group takes much, much longer to go into a complete response. If you are not responding quickly, sometimes doctors change treatment. For that CD2 group, if you know that they are CD2 you can be much more confident about carrying on with the treatment because once these cases go into a response, they stay in a response for longer and actually their long-term clinical outcome is better and so that is really important information. So for people that don’t get CR, they can really worry about that, but if they are in this CD2 group, they should be heavily reassured. Dr. Heuck: Important in that setting is patients who are in the CD2 group who don’t get complete response and all of a sudden get labeled as “resistant” to one drug or the other, which in fact they are not. That is critically important for patients’ outcomes. Jenny: And where does this CD1 or CD2 show up – on the GEP? Dr. Morgan: Yes. It shows 7 molecular groups as well as risk status. Jenny: I have a high-risk feature and haven’t heard about CD1 or CD2, so that is interesting. Dr. Weinhold: There is another important point. High-risk and low-risk patients show comparable response rates – that means many of the high-risk patients reach a complete remission but nevertheless most of them immediately relapse. The idea what we could do in terms of MRD diagnostics would be to combine the GEP70 data collected at baseline, where the MRD resides for some complete remission and just use it for prediction. If a patient has high-risk but reaches complete remission, that may mean that this patient may show immediate relapse. That may be a way to merge GEP data with the MRD test. Jenny: Well, I think the more you know the better off everyone is. Patients have a really tough time trying to make treatment decisions like this friend of mine. It’s a real struggle. Dr. Morgan: It is. Information is power when you are making treatment decisions. Making decisions based on your gut feel is not good and the more information you have that come from these tests, I think the easier it gets to make the correct decision. Jenny: Absolutely. Gary had a follow up question on your point where you were talking about how some treatments may actually ignite aggressive behavior in the clone. He said, “If treatment does excite and energize the previous high-risk clones, this seems like a catch-22. What would be potential treatment options in this case.” I know we are jumping topics pretty significantly. Dr. Morgan: That’s a real question – a proper, informed and scientific question that needs a complicated answer. When we only had chemo, I think the answer to that question was well, ok, it is an important question, but maybe there is nothing we can do about it. But now we don’t just have chemo drugs. We have antibodies and we have targeted treatment and there’s a good example in chronic lymphocytic leukemia. There’s a group with a chromosome deletion 17p which inactivates this gene p53, which means those cases don’t respond to chemotherapy. Now they have this new drug, ibrutinib. People with loss of 17p and p53 inactivation respond beautifully well to this ibrutinib drug. So there immediately, you can see how you make a treatment decision. If you are impacted by p53 you will respond to chemo if you are not, you shouldn’t even be exposed to chemo, you should go straight for a targeted agent. And that’s the way we need to go in multiple myeloma for the high-risk patients. Myeloma patients have del 17p as well and it may be, when we have the information, that we should treat those patients in an entirely different fashion. Jenny: Well this is to me just one more reason to see a multiple myeloma specialist. I advocate that all of the time and so do the other patients because there is just no way a general oncologist can determine all of the nuances that come with this disease. Dr. Morgan: It is remarkably useful. There is also a way of doing it as well. The Signal Genetics test turns around in four days, perhaps a little bit longer for other people. You can get sequencing results back in two weeks. It’s a whole change in the universe really where these tests are a routine part of clinical care and we’re only going to learn how to use them if we actually use them. If you don’t use them, you’ll never get the information. I’m very supportive of this approach. Jenny: Well Cynthia’s additional question is “In your proposal, you said that if you found a dominant clone that was greater than 25% that those patients tended to do better because they might actually keep down the more aggressive clone. Does it matter what that clone type is? (that dominant clone)“ Dr. Morgan: This is entirely a research concept. You have to be careful in a patient because if you have a dominant clone that is aggressive and is dividing and proliferating, you need to kill that clone straightaway because it will just come to dominate everything else. If it’s a clone that gives rise to smoldering myeloma, and we all know that smoldering myeloma can be either low risk or high-risk. If they are a low risk patient and they have a dominant clone, I think that would make you happier just to watch and wait. If they were a high-risk smoldering myeloma patient, you’d probably make a decision to treat if there was a predictive 18 month time where you would develop real myeloma with end organ damage. I think it’s dangerous just to watch and wait. The world of smoldering myeloma is also being changed as a result of these gene expression profiling tests and we developed the GEP to predict people who would do the best. I think its our advice to have the test and if you are high-risk smoldering myeloma to get treated either in a clinical trial or with something that doesn’t include much chemotherapy. Jenny: But in smoldering myeloma isn’t high-risk determined by clinical presentation, like you have X amount of bone damage, or you have this M-protein spike or is more the genetics of the disease. Dr. Morgan: You couldn’t tell. The guys from the Mayo had predictors which included the % of plasma cells in the bone marrow and this thing called the light chain ratio. The Spanish had a flow cytometry test where you could look at the surface of the plasma cells and the normal cells. But again in this situation, a slightly different GEP profile called GEP 4 outperforms those other two tests and identifies an 80-90% chance of progressing to myeloma that needs treatment within a 12-18 month period. I think that is very important as well. It helps you make a sound decision about watching and waiting or intervening early. Jenny: We have a lot of smoldering myeloma listeners, so that would be very important for them. My last question in general – we ask this question of everyone. Can you explain the milestones that you have in mind and just an estimated cost of what you think it would take to develop what you need to develop to move forward? Dr. Weinhold: We planned four milestones. Milestone 1 would be the generation of disease avatars for 10 high-risk patients. We plan to start on month 1 and finish on month 3. The estimated costs would be approximately $40,000. Milestone 2 would be molecular analyzers of pre-engraftment samples from these patients but also samples from untreated disease avatars. From these 10 avatars that we have developed for these 10 patients, it will include whole exome sequencing that means we will determine the sequence of all genes in these high-risk tumors but we will also determine the expression levels of these genes using a modern method called IM86 – also called next generation sequencing method developed to analyze expression levels. The idea is to start this sub-project in month 7 and finish in month 10. The estimate costs would be significantly higher just because of the inclusion of these modern methods and would be approximately $90,000. Milestone 3 would be to treat these 10 with bortezomib. We would start in week 3 and we would plan to finish it in week 22 and the approximate costs would be $30,000. The last milestone, milestone 4, would be to perform molecular analyzers of treated high-risk avatars. We would start treating 10 of them but we expect that what we get in the first year of funding is sufficient for molecular analyzers for 5 of these treated avatars. We would start with this in month 8 and plan to finish it at the end of year 1 and the estimated cost were approximately $40,000. And again we will perform whole exome and RNA sequencing. That’s our plan. Jenny: That’s great. That sounds like you have it very well defined which is wonderful. To me and to the other patients listening to this, this project is significant because it allows you to shorten up the standard clinical trial process. So instead of sub-typing patients by genetic mutation, or by clonal presentation, and then trying to craft a clinical trial around those patients with one particular drug you are going to test in that clinical trial, this allows you to completely change the way you construct a possible clinical trial and just shorten it up by decades, it seems to me. Dr. Weinhold: That is a good assessment, yes. That is really what we would like to do. Jenny: We love what you are doing. Dr. Morgan, Dr. Weinhold, Dr. Heuck, Dr. Epstein and Dr. Yaccoby, we are so appreciative for you joining us today. What you are doing is very significant and we really appreciate your time. Drs: Thank you for the opportunity. Jenny: Thank you for listening to Myeloma Crowd Radio and the new MCRI series. Patients can support the discovery of a cure and we encourage you to become involved.
about the author
Myeloma survivor, patient advocate, wife, mom of 6. Believer that patients can help accelerate a cure by weighing in and participating in clinical research. Founder of Myeloma Crowd by HealthTree and the HealthTree Foundation.